Abstract Architecture

F-ONE process digitalization approach compared to
using an RPA platform for process automation

Case study

Some limitations of RPA platforms

You have spoken to a sales account executive from a well-known Robotic Process Automation (RPA) vendor and are excited about the opportunities thy presented. Apparently, RPA bots can be developed by ordinary business users, can automate any task that is repetitive and rule-based, and they even include AI capabilities to automate cognitive tasks. You run a Proof of Concept project and indeed, within a few weeks the robot is up and running.

The problems surface later, when you realize that you cannot trust your robotic workforce with business-critical processes. The robots are very sensitive to changes in the UI and require regular code maintenance. They lack memory, so that when they break down, the best you get is an error report to deal with a failed process instance. You set up a dedicated Center of Excellence and trained people in operating the platform, but finding qualified developers is not easy, and it slows down your automation drive. Most importantly, your RPA suite is not very good at coordinating multiple actors such as people, production equipment, IOT endpoints, and external stakeholders, which would be critical for end-to-end process automation.  

Robotic Process Automation (RPA) vendors such as UiPath, Automation Anywhere and Blue Prism have become very prominent in the process automation world, thanks to the ubiquity of potential RPA applications and to their mighty marketing budgets. 

However, it is important to understand process automation actually comprises two important components: control flow automation (what happens after each process step) and task automation (how each process step is performed). 


RPA, which means emulating human interactions with IT systems, is an approach to task automation. Thus, RPA suites originally lacked control flow automation capabilities. While most vendors have added control flow automation features, those features are still centered around using RPA, which makes them much less powerful and versatile compared to specialized control flow automation solutions such as workflow engines. 

F-ONE way of process automation transcends those limitations

Control flow is automated using a specialized solution (workflow engine), which has powerful capabilities to deal with exceptions, approvals, long-running processes, external interfaces, databases, IOT equipment etc. It is also open-source, effectively eliminating vendor dependency. 

The logic of process automation is laid bare in a BPMN diagram, which is the accepted industry standard for complex process automation. You can use one of the great enterprise-grade tools to create and collaborate on process models. 

You can still use your RPA platform of choice to automation UI-based interactions, or you can use a powerful Python- or Java-based RPA library of your choosing. In addition, you can seamlessly integrate APIs, microservices, human workflows and external partners to achieve end-to-end process automation. 

Thus the automations are more transparent and powerful, less expensive, require less maintenance, remain flexible over time, avoid introducing technical debt to your organization. 

We are here to help you

We have spent much time discovering the best practices for future-proof process digitalization. Please get in touch to discuss your ideas and requirements and we will be happy to support you on your process digitalization journey.